Diabetes affects over 25 million Americans; 50% have distal symmetric polyneuropathy (DSP). DSP is a leading cause of disability and reduced quality of life. Once established, it is difficult to reverse. Early DSP progresses slowly and currently available surrogate endpoints do not change significantly. Early DSP is characterized by progressive loss of small unmyelinated nerve fibers. Skin biopsy with measurement of intraepidermal nerve fiber density (IENFD) is a reproducible tool to assess these fibers. Data from the Utah Diabetic Neuropathy Study indicate IENFD declines prior to DSP onset. Patients who experience progressive neuropathy have a significantly lower baseline IENFD. These findings suggest IENFD is a sensitive diagnostic measure that could be used to evaluate treatment efficacy. However, IENFD is invasive, technically demanding, expensive, and inconvenient. Corneal confocal microscopy (CCM) noninvasively and directly visualizes small unmyelinated axons on the cornea. CCM is well tolerated and rapid, and is a promising alternative to IENFD. Because reversible metabolic injury segues into irreversible axon loss, there is consensus that therapy should be initiated early. An ideal strategy is identification patients at high DSP risk using a screening strategy in order to facilitate prevention or early intervention. Annual screening for diabetic retinopathy has reduced the risk of diabetes related blindness and serves as a model. Availability of a similar screening method for DSP would enhance patient care and the ability to evaluate novel therapeutics. The proposed studies will evaluate CCM as a diagnostic and screening tool, and surrogate progression measure. There are a number of critical issues that must be resolved before CCM can be routinely used as a diagnostic test or surrogate measure, including demonstration of its clinical meaning and responsiveness to change. Because DSP is a length dependent process, the observation of abnormal corneal innervation is unexpected. Rigorous prospective evaluation is therefore necessary to establish utility. Preliminary data indicate CCM estimation of nerve fiber length (NFL) is highly reproducible, well tolerated and efficient and that NFL and other measures are reduced diabetic patients, more so in those with DSP. Specific aims include development of age and sex stratified normal data, determination of CCM's diagnostic utility and its responsiveness to DSP progression. The clinical meaning of CCM will be assessed by correlation with validated patient reported neuropathy questionnaires and functional measures of mobility and balance. These aims will be achieved by recruiting patients scheduled for their annual retinopathy screening. A major goal is to determine if CCM could be used as an effective screening strategy that would take place at the same time and location as yearly retinal screening. Early identification of patient with DSP would allow clinicians to begin treatment (e.g. lifestyle or ris modification strategies) while DSP is potentially reversible. Development of simple predictive tools for DSP risk would facilitate practically achievable prevention studies, something impossible using current methodologies.